Hepatology highlights


  • Potential conflict of interest: Nothing to report.

Sniffing Out the Cause of PBC

The study by Ala et al. gives the hounds real hope as they corner their quarry. In New York, cluster analysis of patients with primary biliary cirrhosis (PBC) according to post-code, showed a significant association with proximity to superfund toxic waste sites (SFS) when these were dumps and landfill sites rather than buildings. With the exception of a “false positive” cluster attributed to physician referral bias adjacent to the study centre at Mount Sinai the overall pattern was consistent, although not all clusters achieved statistical significance. Clustering was shown in two sets; those listed for liver transplantation between 1995 -2003 (n=73) and a separate cohort of 172 non-listed patients entered on a clinical research database between 1990-2004 (see Fig.). No such clustering occurred with primary sclerosing cholangitis (PSC) which acted as a control cholestatic in chronic liver disease. The SFS sites were contaminated with benzene alone or both benzene and other organic compounds, including chlorinated hydrocarbons. The findings are compatible with a role for inhalation of volatile toxins or airborne particle bound chlorinated hydrocarbons in the pathogenesis of PBC. The literature already contains experimental data outlining how benzene and its derivatives have a predilection for hepatotoxicity in the female and how, in halogenated forms, they may mimic the lipoylated autoantigen of pyruvate dehydrogenase and thus induce the formation of the diagnostic M2 antimitochondrial antibody. There is already overwhelming proof from previous reports of a genetic predisposition to PBC and a significant contribution to risk from gender. The simplest and therefore most appealing hypothesis must be of a genetically determined weakness in the liver's ability to safely detoxify and eliminate hazardous xenobiotics, exposure to which has geographical location among its determinants. (See HEPATOLOGY 2006;43:525-531).

Illustration 1.

Rapid Relief From Pruritus in BRIC—Can Cholestasis Be Hypercholeretic?

Benign recurrent intrahepatic cholestasis (BRIC) may cause many attacks of severe pruritus and jaundice in predisposed individuals. Stapelbroek et al. report on 3 patients of their own in whom medical treatment with cholestyramine, ursodeoxycholic acid, antihistamines, rifampicin, naltrexone, prednisolone and ultraviolet light had failed. In all 3 patients pruritus resolved within 24 hours of establishing nasobiliary drainage (NBD). If this appears counterintuitive, the reader can at least be reassured that in two instances (2 of 9 in a worldwide experience quoted) where NBD yielded no bile, no relief occurred. The reported patients had defined mutations in ATP8B1, encoding FIC1, a putative phospholipid flippase. Patients responding to the anticholestatic agents listed above are more likely to have mutations in ABCB11, encoding BSEP the bile salt transporter (BRIC2). In BRIC1 which is associated with mutations of FIC1 and in accordance with an animal model, patients appear to have an excessive flux of bile acids through their liver with an imbalance of accompanying phospholipids. Enrichment of the phospholipids within the luminal surface of the bile canalicular membrane with phosphatidyl choline is likely to be a major function of the FIC1 transporter. In BRIC1 biliary concentrations of phosphatidyl choline were reduced and those of lyso-phosphatidyl choline and shingomyelin increased in proportion. It appears, therefore, that the attack of BRIC1 was associated with the liver's inability to keep up with the demand for biliary phospholipid, and in particular the requirement for enrichment of the luminal canalicular membrane with phosphatidyl choline. NBD, by interruption of the enterohepatic circulation, has provided the liver with temporary respite from overwhelming bile acid flux and sufficient time to replenish its stock of appropriately positioned phospholipid. (See HEPATOLOGY 2006;43:51-53.)

Adenomas and Their Potential for Malignant Transformation

A consensus exists among hepatologists which favors resection of hepatic adenoma as prophylaxis against bleeding prior to any pregnancy and whenever else the risk of bleeding is deemed appreciable. Adenoma resection solely on the basis of removing a potential for developing hepatocellular carcinoma (HCC) is not always as easy to justify in terms of risk-benefit ratio, especially if major surgery is required. The article by Zucmann-Rossi et al. helps to refine the classification of adenoma according to its genotype and histology and may help to inform the above decision making. Of a series of 96 adenomas, 44 were associated with mutations in hepatocyte nuclear factor 1α (HNF1α) and were characterized by marked steatosis and a lack of cytological abnormalities, although 2 cases were deemed borderline between adenoma and HCC and in 1 case part of the adenoma was associated with an HCC. Mutations of β-catenin were found in a separate series of 12 adenomas including 2 which were deemed borderline between adenomas and HCC, 3 in which adenoma was associated with HCC and 1 which was reclassified as HCC. The authors propose a classification of hepatic adenomas based on three criteria, HNF1α mutations, β-catenin activation and the presence or absence of inflammatory infiltrates. Only 5 of the 96 lesions had combined characteristics which defied allotment to one of these 4 groups. The Figure presents the results in summary form, reflecting the frequency of each type and its relative malignant potential. The study has great merit as a basis both for future clinical evaluation of patients and prospective research into the biological behavior of these vexing lesions. (See HEPATOLOGY 2006;43:515-524.)

Illustration 2.

Polymorphisms in Hepatic Metabolism of Alcohol May Promote Alcoholism But Not Alcoholic Liver Disease

It is well documented in East Asians that possession of a poorly functioning form of acetaldehyde dehydrogenase (ALDH2) protects against alcoholism. Those possessing this faulty variant, who nevertheless are not deterred by the noxious effects of acetaldehyde from becoming alcoholic, probably have an increased susceptibility to develop alcoholic liver disease. Key hepatic enzymes involved in metabolism of ethanol and acetaldehyde are similarly affected by single nucleotide polymorphisms which alter their functional ability. This knowledge has prompted many studies to seek similar associations in other racial groups. Numerous studies have searched for associations of polymorphisms of alcohol dehydrogenase (ADH), cytochrome P450 2E1(CYP2E1), and acetaldehyde dehydrogenase (ALDH) with alcoholism and/or alcoholic liver disease. Zintzaras et al. performed a metaanalyis of 50 such studies. Positive associations between enzyme polymorphisms of ADH and ALDH and the risk of alcoholism per se were confirmed, mainly when analysis was confined to specific racial subgroups. There was no detectable influence of the CYP2E1 polymorphism. No association was identified between any polymorphic enzyme variant and the risk of developing alcoholic liver disease. Therefore, in the search for the basis for variability between individuals in their susceptibility to alcoholic liver disease only minor significance can be ascribed directly to the pharmacokinetics of ethanol and acetaldehyde metabolism. (See HEPATOLOGY 2006;43:352-361.)

Cerebral Ammonia Uptake in Hepatic Encephalopathy — It's Simply a Matter of Diffusion

Positron Emission Tomography (PET) scanning was used to measure brain ammonia (13NH 3) uptake in the first such study to enroll patients with clinically overt hepatic encephalopathy. Keiding et al. calculated a blood flow–independent permeability-surface area product (PS) to characterize net transfer of 13NH 3 from blood to brain across the blood-brain barrier (PSBBB) and measured its conversion to 13N-glutamine. The PSBBB for cerebral 13NH 3 uptake was markedly reduced in HE Grades II-IV (mean 0.15mL/min/mL) compared with grade I-II HE (0.26). Cerebral blood flow (CBF) was measured by dynamic 15O-water scans and in comparison with controls revealed reduced CBF in cirrhosis. Further reduction in CBF was seen in those with more severe grades of HE as well as regional differences, which applied in all groups, whereby cortical CBF was only two thirds of that in the basal ganglia and cerebellum. Kinetics of conversion of 13NH 3 to glutamine (PSmet) were not significantly altered in cirrhosis with HE compared to controls. There was a non significant tendency for higher PSmet in basal ganglia compared to cortex and cerebellum which applied to controls and cirrhotic patients with mild or severe HE. During the 30-minute period of these studies no efflux of cerebral 13N-glutamine could be detected. The main determinant of brain ammonia uptake was clearly blood ammonia concentration (see Fig.) with passive diffusion its uptake mechanism. No support was found for an effect of extracellular metabolic alkalosis on cerebral NH4+ uptake as is widely supposed to occur. The study points clearly to the blood ammonia concentration being the critical driving force behind increased cerebral concentrations of ammonia and glutamine in HE. These data therefore strongly support the logic behind attempts to reduce blood ammonia concentrations as a key preventive measure in prophylaxis against HE. (See HEPATOLOGY 2006;43:42-50.)

Illustration 3.

Age-Dependent Variants of AIH

Autoimmune hepatitis has historically been thought of as a disease of young females characterized by autoantibodies, hyperglobulinemia and responsiveness to treatment with corticosteroids. Identification of different autoantibodies has led to subclassification into subtypes. The classical AIH patient with anti smooth muscle antibody, anti nuclear antibody and elevated serum IgG represents Type I autoimmune hepatitis; those with liver kidney microsomal (LKM) antibodies, Type II. The article by Czaja et al. invites further subclassification of Type I AIH based on age distribution, HLA frequencies, associated autoimmune disorders and responsiveness to corticosteroid treatment. In comparing cohorts of patients aged <30 (n = 37) and >60 years (n = 47) there were contrariwise differences in the incidence of HLA frequencies with HLA DR3 being relatively less frequent (23% vs. 58% P = .004); and HLA DR4 relatively more frequent in the elderly (47% vs. 13%; P = .003). Those <30 years showed a lower incidence of cirrhosis at presentation (10% vs. 33% P = .03) and of thyroid and rheumatoid disease (13 vs. 42% P = .006). This may be attributed, at least in part, to a cohort effect with accumulation of co-morbidity and advanced disease with ageing, although, the reverse trend exists for ulcerative colitis and autoimmune hemolysis (13% % vs. 0%. P = .02). Failure to respond to corticosteroid treatment (24% vs. 5%; P = .03) was also more common in the younger cohort. (See HEPATOLOGY 2006;43:532-538.)